Fix multiple parsing bugs and add RFC 4180 quote support

- Change nread to size_t to prevent overflow.

- Fix state machine to handle consecutive newlines appropriately.

- Add csv_parser_finish to emit pending trailing fields at EOF.

- Fix nread boundary count on callback errors.

- Add support for quoted fields, including escaped quotes.

- Update and expand test suite to cover new cases and quote parsing.

3 files changed, 176 insertions, 27 deletions

diff --git a/src/csv_parser.c b/src/csv_parser.c
index b93713a..32cf8d4 100644
--- a/src/csv_parser.c
+++ b/src/csv_parser.c
@@ -70,8 +70,15 @@ size_t csv_parser_execute(csv_parser_t *parser,
                 break;
             case csvps_field_start:
                 parser->col += 1;
-                parser->state = csvps_field_value;
-                field_value = cursor;
+                if (ch == '"') {
+                    parser->state = csvps_field_quoted_value;
+                    cursor++;
+                    parser->nread++;
+                    field_value = cursor;
+                } else {
+                    parser->state = csvps_field_value;
+                    field_value = cursor;
+                }
                 break;
             case csvps_field_value:
                 if (ch == settings->delimiter) {
@@ -79,10 +86,6 @@ size_t csv_parser_execute(csv_parser_t *parser,
                 } else if (ch == '\r' || ch == '\n') {
                     parser->state = csvps_line_end_begin;
                 } else {
-                    // If we previously been in csvps_field_value state
-                    // and field_value is not set, set it right away.
-                    // This can happen when we execute parser multiple
-                    // times in the value state.
                     if (field_value == NULL) {
                         field_value = cursor;
                     }
@@ -97,6 +100,7 @@ size_t csv_parser_execute(csv_parser_t *parser,
                                                    parser->col);
                             if (r) {
                                 parser->state = csvps_error;
+                                parser->nread++;
                                 return parser->nread;
                             }
                         }
@@ -104,6 +108,67 @@ size_t csv_parser_execute(csv_parser_t *parser,
                     parser->nread++;
                 }
                 break;
+            case csvps_field_quoted_value:
+                if (field_value == NULL) {
+                    field_value = cursor;
+                }
+                if (ch == '"') {
+                    parser->state = csvps_field_quoted_quote;
+                    if (settings->field_cb && field_value && (cursor >= field_value)) {
+                        r = settings->field_cb(parser,
+                                               field_value,
+                                               cursor - field_value,
+                                               parser->row,
+                                               parser->col);
+                        if (r) {
+                            parser->state = csvps_error;
+                            return parser->nread;
+                        }
+                    }
+                    field_value = NULL;
+                    cursor++;
+                    parser->nread++;
+                } else {
+                    cursor++;
+                    if (cursor == data_end) {
+                        if (settings->field_cb && field_value) {
+                            r = settings->field_cb(parser,
+                                                   field_value,
+                                                   cursor - field_value,
+                                                   parser->row,
+                                                   parser->col);
+                            if (r) {
+                                parser->state = csvps_error;
+                                parser->nread++;
+                                return parser->nread;
+                            }
+                        }
+                    }
+                    parser->nread++;
+                }
+                break;
+            case csvps_field_quoted_quote:
+                if (ch == '"') {
+                    if (settings->field_cb) {
+                        r = settings->field_cb(parser, cursor, 1, parser->row, parser->col);
+                        if (r) {
+                            parser->state = csvps_error;
+                            return parser->nread;
+                        }
+                    }
+                    parser->state = csvps_field_quoted_value;
+                    cursor++;
+                    parser->nread++;
+                    field_value = cursor;
+                } else if (ch == settings->delimiter) {
+                    parser->state = csvps_field_end;
+                } else if (ch == '\r' || ch == '\n') {
+                    parser->state = csvps_line_end_begin;
+                } else {
+                    parser->state = csvps_error;
+                    return parser->nread;
+                }
+                break;
             case csvps_field_end:
                 // callback
                 if (settings->field_cb && field_value) {
@@ -135,15 +200,20 @@ size_t csv_parser_execute(csv_parser_t *parser,
                         return parser->nread;
                     }
                 }
-                parser->state = csvps_line_end;
+                if (ch == '\r') {
+                    parser->state = csvps_line_end;
+                } else {
+                    parser->state = csvps_line_start;
+                }
+                cursor++;
+                parser->nread++;
                 break;
             case csvps_line_end:
-                if (ch == '\r' || ch == '\n') {
+                if (ch == '\n') {
                     cursor++;
                     parser->nread++;
-                } else {
-                    parser->state = csvps_line_start;
                 }
+                parser->state = csvps_line_start;
                 break;
             case csvps_error:
                 return parser->nread;
@@ -156,3 +226,21 @@ size_t csv_parser_execute(csv_parser_t *parser,
 
     return parser->nread;
 }
+
+int csv_parser_finish(csv_parser_t *parser, const csv_parser_settings_t *settings)
+{
+    int r = 0;
+    if (parser->state == csvps_field_start) {
+        if (settings->field_cb) {
+            parser->col += 1;
+            r = settings->field_cb(parser, "", 0, parser->row, parser->col);
+            if (r) {
+                parser->state = csvps_error;
+            }
+        }
+    }
+    if (parser->state != csvps_error) {
+        parser->state = csvps_line_start;
+    }
+    return r;
+}
diff --git a/src/csv_parser.h b/src/csv_parser.h
index 8c78e8b..267c768 100644
--- a/src/csv_parser.h
+++ b/src/csv_parser.h
@@ -35,6 +35,8 @@ typedef enum csv_parser_state_t {
     csvps_line_start = 0,
     csvps_field_start,
     csvps_field_value,
+    csvps_field_quoted_value,
+    csvps_field_quoted_quote,
     csvps_field_end,
     csvps_line_end_begin,
     csvps_line_end,
@@ -48,7 +50,7 @@ typedef struct csv_parser_t {
     int row;
     /* csv column */
     int col;
-    uint32_t nread;
+    size_t nread;
     /* user data */
     void *data;
 } csv_parser_t;
@@ -88,6 +90,12 @@ size_t csv_parser_execute(csv_parser_t *parser,
                           const char *data,
                           size_t data_len);
 
+/**
+ * Flush any pending state. This is useful when the file ends
+ * without a trailing newline, to flush the final empty field.
+ * Returns 0 on success, anything else on error.
+ */
+int csv_parser_finish(csv_parser_t *parser, const csv_parser_settings_t *settings);
 
 #ifdef __cplusplus
 }
diff --git a/tests/main.c b/tests/main.c
index 90863d8..00a7add 100644
--- a/tests/main.c
+++ b/tests/main.c
@@ -11,44 +11,83 @@ struct test_results {
 };
 
 
-// 5 items per row, 3 rows per test case: 15 * 6 slots
+// 5 items per row, 4 rows per test case: 20 slots
 static int expected_test_results[] = {
-    // ""
+    // 0: ""
     -1, -1, -1, -1, -1,
     -1, -1, -1, -1, -1,
     -1, -1, -1, -1, -1,
-    // ", "
+    -1, -1, -1, -1, -1,
+    // 1: ", "
     0, 1, -1, -1, -1,
     -1, -1, -1, -1, -1,
     -1, -1, -1, -1, -1,
-    // ","
-    0, -1, -1, -1, -1,
+    -1, -1, -1, -1, -1,
+    // 2: ","
+    0, 0, -1, -1, -1,
+    -1, -1, -1, -1, -1,
     -1, -1, -1, -1, -1,
     -1, -1, -1, -1, -1,
-    // "\n"
+    // 3: "\n"
     -1, -1, -1, -1, -1,
     -1, -1, -1, -1, -1,
     -1, -1, -1, -1, -1,
-    // "\r\n",
     -1, -1, -1, -1, -1,
+    // 4: "\r\n"
     -1, -1, -1, -1, -1,
     -1, -1, -1, -1, -1,
-    // "a,b,,c,"
-    1, 2, 0, 1, -1,
     -1, -1, -1, -1, -1,
     -1, -1, -1, -1, -1,
-    // "a,b,,c, "
+    // 5: "a,bb,,c,"
+    1, 2, 0, 1, 0,
+    -1, -1, -1, -1, -1,
+    -1, -1, -1, -1, -1,
+    -1, -1, -1, -1, -1,
+    // 6: "a,bb,,c, "
     1, 2, 0, 1, 1,
     -1, -1, -1, -1, -1,
     -1, -1, -1, -1, -1,
-    // "a,b,\na,b,"
+    -1, -1, -1, -1, -1,
+    // 7: "a,bb,\na,b,"
     1, 2, 0, -1, -1,
-    1, 1, -1, -1, -1,
+    1, 1, 0, -1, -1,
     -1, -1, -1, -1, -1,
-    // " bc "
+    -1, -1, -1, -1, -1,
+    // 8: " bc "
     4, -1, -1, -1, -1,
     -1, -1, -1, -1, -1,
     -1, -1, -1, -1, -1,
+    -1, -1, -1, -1, -1,
+    // 9: "a\n\nb" (empty line in middle)
+    1, -1, -1, -1, -1,
+    -1, -1, -1, -1, -1,
+    1, -1, -1, -1, -1,
+    -1, -1, -1, -1, -1,
+    // 10: "\"a\"" (quotes)
+    1, -1, -1, -1, -1,
+    -1, -1, -1, -1, -1,
+    -1, -1, -1, -1, -1,
+    -1, -1, -1, -1, -1,
+    // 11: "\"a,b\"" (quotes with delimiter)
+    3, -1, -1, -1, -1,
+    -1, -1, -1, -1, -1,
+    -1, -1, -1, -1, -1,
+    -1, -1, -1, -1, -1,
+    // 12: "\"a\nb\"" (quotes with newline)
+    3, -1, -1, -1, -1,
+    -1, -1, -1, -1, -1,
+    -1, -1, -1, -1, -1,
+    -1, -1, -1, -1, -1,
+    // 13: "\"a\"\"b\"" (escaped quotes)
+    3, -1, -1, -1, -1,
+    -1, -1, -1, -1, -1,
+    -1, -1, -1, -1, -1,
+    -1, -1, -1, -1, -1,
+    // 14: "\"\"" (empty quotes)
+    0, -1, -1, -1, -1,
+    -1, -1, -1, -1, -1,
+    -1, -1, -1, -1, -1,
+    -1, -1, -1, -1, -1,
 };
 
 
@@ -66,6 +105,12 @@ static const char *test_data[] = {
     "a,bb,,c, ",
     "a,bb,\na,b,",
     " bc ",
+    "a\n\nb",
+    "\"a\"",
+    "\"a,b\"",
+    "\"a\nb\"",
+    "\"a\"\"b\"",
+    "\"\""
 };
 
 
@@ -73,12 +118,18 @@ static int test_data_end_states[] = {
     csvps_line_start,
     csvps_field_value,
     csvps_field_start,
-    csvps_line_end,
-    csvps_line_end,
+    csvps_line_start,
+    csvps_line_start,
     csvps_field_start,
     csvps_field_value,
     csvps_field_start,
     csvps_field_value,
+    csvps_field_value,
+    csvps_field_quoted_quote,
+    csvps_field_quoted_quote,
+    csvps_field_quoted_quote,
+    csvps_field_quoted_quote,
+    csvps_field_quoted_quote,
 };
 
 static int num_test_cases = sizeof(test_data)/(sizeof(char*));
@@ -110,7 +161,7 @@ void print_results(int *data, int len)
 int field_cb(csv_parser_t *parser, const char *data, size_t length, int row, int col)
 {
     struct test_results *results = parser->data;
-    int slot = results->test_case_number * 15 + row * 5 + col;
+    int slot = results->test_case_number * 20 + row * 5 + col;
 
     if (results->results[slot] == -1) {
         results->results[slot] = length;
@@ -182,6 +233,7 @@ START_TEST(test_parser_full)
         int nread = csv_parser_execute(&parser, &settings, str, len);
         ck_assert_int_eq(len, nread);
         ck_assert_int_eq(parser.state, test_data_end_states[i]);
+        csv_parser_finish(&parser, &settings);
     }
 
     printf("Test results:\n");
@@ -221,6 +273,7 @@ START_TEST(test_parser_chunked)
             ck_assert_int_eq(1, nread);
         }
         ck_assert_int_eq(parser.state, test_data_end_states[i]);
+        csv_parser_finish(&parser, &settings);
     }
 
     printf("Test results:\n");